JP5675825B2 - Multi-channel nonlinearity compensation in optical communication links - Google Patents

Description

The present invention relates generally to optical communication systems, and more particularly to a method and apparatus for compensation of interchannel nonlinear distortion in a wavelength division multiplexing (WDM) transmission system.

BACKGROUND OF THE INVENTION Modern optical communication systems include network nodes interconnected by optical waveguides, which are typically single mode optical fiber links. Within the network node, communication signals are converted between an electrical format used for signal processing, including playback and routing, and an optical format used for transmission between nodes. The links between nodes typically have multiple connected optical components, each containing multiple optical fiber spans that are tens of kilometers in length, and the attenuation experienced by the optical signal during transmission through those fiber spans. And a corresponding optical amplifier.

  Signals transmitted over optical fiber links are affected by linear dispersion processes such as chromatic dispersion and polarization mode dispersion. In particular, chromatic dispersion may be compensated at any convenient point within the transmission system, and many existing installed fiber optic transmission links are installed at network nodes and / or amplifier locations, such as dispersion compensating fiber length. Includes compensation components. The technique of deploying in-line dispersion compensation components in an optical transmission system is sometimes referred to as “dispersion management”, and the resulting cumulative dispersion characteristic in the transmission link is referred to as a “dispersion map”.

  While distributed management techniques can be very effective, they also have the disadvantage of lacking flexibility. For example, an effective dispersion map for a specific wavelength channel within a specific transmission link, for a channel transmitted at a different wavelength, and / or all or part of the link later incorporated into a different optical transmission line In some cases, the effect may be low. Therefore, upgrading a dispersion-managed link or reconfiguring the optical network in which the link is installed to carry multiple wavelength division multiplexing (WDM) channels is a redesign of the dispersion management strategy. , And dispersion compensation components in the link may need to be changed. Greater flexibility may be obtained using an electronic dispersion compensation method such as that described in US patent application Ser. No. 12/085571 (also published as international application WO2007 / 041799), This patent application specifically discloses an electronic dispersion compensation method that uses block coding of information and frequency domain equalization of the resulting received signal to provide complete compensation of linear dispersion in the electrical domain. ing. This approach can be implemented particularly advantageously if an orthogonal frequency division multiplexing (OFDM) method is used for encoding and decoding the electrical signal. Nevertheless, many existing installed transmission systems use fixed distributed management.

  In addition to linear processes, the propagation of optical signals can also suffer from non-linear effects. Although the level of optical nonlinearity present in most practical transmission media, especially silica-based optical fibers, is relatively low, in order to maintain a sufficient signal-to-noise ratio over long transmission distances, It is desirable to transmit optical signals at high power levels. The use of high transmit power increases the effects of optical non-linearity and results in optical signal distortion, ultimately limiting the received signal quality, so that the signal can be detected, recovered, and replayed It will limit the maximum achievable transmission distance. Therefore, it is desirable to mitigate as much as possible the effects of nonlinear distortion and the effects of linear processes such as dispersion. US patent application Ser. No. 12 / 445,386 (also published as international application WO 2008/074085) describes signal processing at the transmitting end (ie before compensation) and / or at the receiving end (ie after compensation). Discloses a method and apparatus for compensating for the effects of optical nonlinearities. This prior art disclosure is generally directed to compensating for the effects of so-called single channels, such as self-phase modulation (SPM), resulting from high transmit power levels in specific channels that cause nonlinear distortion in the channels themselves. . However, in WDM systems where information is transmitted using many different wavelength channels, the contribution of a single channel to the overall optical power is relatively small, such as four-wave mixing (FWM) or cross-phase modulation (XPM). So-called channel-to-channel effects can be important.

  It is believed that in-line dispersion compensation, such as dispersion management techniques used in many existing optical transmission links, can increase the level of nonlinear distortion. This increase in distortion is caused by the enhancement of nonlinear mixing between WDM channels, such as with XPM. This is because the “walk-off” between WDM channels is reduced by using distributed management. In particular, dispersion propagates optical fibers through different WDM channels at different speeds, so the “averaged” effect reduces the severity of nonlinear distortion. This is because no part of the signal is placed continuously under the influence of a specific peak in the overall optical power. However, this benefit is somewhat reduced when in-line dispersion compensation, ie dispersion management, is present.

  In general, it is considered that interchannel effects such as XPM cannot be effectively mitigated by electronic processing. This is because to nullify the effects of nonlinear propagation requires equalization based on the complete representation of all channels transmitted in the fiber and their propagation characteristics. On the other hand, as described above, the use of optical compensation techniques generally increases the flexibility of the network, particularly in connection with performing upfade and / or reconfiguration in a simple and cost effective manner. Limited.

US Patent Application No. 12/085571 (International Application WO2007 / 041799) U.S. Patent Application No. 12/445386 (International Application WO2008 / 074085)

  Therefore, the object of the present invention can effectively and efficiently apply compensation and / or mitigation of non-linear distortion, especially inter-channel effects, to existing transmission links including transmission links using dispersion management strategies, It is also provided in such a way that it can also be adapted to support network upgrades and reconfigurations.

SUMMARY OF THE INVENTION In one aspect, the present invention is a method for mitigating interchannel nonlinear distortion of an optical signal carried on one of a plurality of wavelength channels in a wavelength division multiplexing (WDM) transmission system, comprising:
Obtaining a measure of the total optical power of a plurality of wavelength channels;
Applying a phase modulation to the optical signal in proportion to the measure of the total optical power.

  As mentioned above, in general, channel-to-channel distortions cannot be effectively mitigated by electronic processing without fully knowing all WDM channels transmitted over the optical link and the associated propagation characteristics. It is done. However, the inventors of the present invention have discovered that a single measure of total optical power across all relevant wavelength channels can be used to achieve a surprisingly effective relaxation of interchannel nonlinear distortion. This is believed to be due to the effects of chromatic dispersion that causes a reduction in the efficiency of interchannel effects such as XPM, both in the modulation frequency and frequency spacing of WDM channels, even in dispersion managed systems. For example, it has been calculated that for a 50 GHz WDM channel spacing, only the components of the signal having a frequency less than 1 GHz can impose a substantial XPM penalty on adjacent channels. This bandwidth is expected to decrease further in the more distant WDM channels.

  Non-linear distortion mitigation may be performed at the receiving end of the optical transmission link. Alternatively or in addition, in-line nonlinear distortion mitigation may be performed at one or more locations in the optical link or network, such as amplifier locations or add / drop nodes.

  Thus, advantageously, a single measure of total optical power can be used as a basis for mitigating interchannel nonlinear distortion. In preferred embodiments, a single photodiode can be used to detect optical power in multiple WDM channels, eg, covering several WDM bands, one or more per WDM channel, etc. It is also possible to use multiple photodiodes to obtain an appropriate measure of total optical power.

  In light of the above consideration, the measure of total optical power is preferably a bandwidth limit measure of instantaneous optical power in a plurality of wavelength channels. As will be appreciated, the use of actual optoelectronic components and electronic components essentially results in a bandwidth limited measurement of the detected optical power. However, in a preferred embodiment, further improvements in mitigating non-channel non-linear distortion can be achieved by adapting this measurement to the system and the specific optical signal.

  More specifically, the bandwidth of the instantaneous optical power measure is preferably limited according to the low pass characteristics. Advantageously, the low-pass characteristic has a bandwidth that is narrower than the bandwidth of the optical signal. More preferably, the low-pass characteristics are selected or optimized to maximize the compensation level of the inter-channel nonlinear distortion of the optical signal. In some embodiments, the use of adaptive digital filters and / or analog filters advantageously allows for dynamic optimization of filter characteristics to facilitate flexible deployment and reconfiguration.

  In the preferred embodiment utilizing a conventional single mode fiber with corresponding nonlinear propagation characteristics, the applied phase modulation is to advance the phase.

  Further, in a preferred embodiment, a proportionality constant between the measure of total optical power and the level of phase modulation is determined according to a measure of the effective nonlinear length of the WDM transmission system.

  Advantageously, the plurality of wavelength channels consists of a band of channels selected from among the more transmission WDM channels. In particular, the optical signal to which the compensation is applied can be located near the center of the band of the selected channel. As mentioned previously, the effects of interchannel nonlinear distortion are greatly reduced in more distant wavelength channels, and therefore it may be desirable to avoid detection of channels that do not substantially contribute to the distortion of the optical signal of interest. is there. The bandwidth of the wavelength channel is encompassed by, for example, a bandwidth of at least 100 GHz, preferably a bandwidth of at least 200 GHz, more preferably a bandwidth of at least 300 GHz.

  In multi-channel systems where nonlinear distortion should be mitigated in multiple optical signals, it may be beneficial to use a different compensation signal (ie, a measure of total optical power) for each optical signal. For example, the optimal bandwidth characteristics of the instantaneous optical power measure can be different for each optical signal. However, since this approach requires additional components and processing, there is a corresponding trade-off between maximizing compensation across all optical signals and cost / complexity. Advantageously, effective mitigation of nonlinear distortion can be achieved without optimization for each individual optical signal, and thus, in some embodiments, a single measure of total optical power in multiple wavelength channels. However, it has been found that it can be used to apply phase modulation to multiple optical signals to mitigate interchannel nonlinear distortion. Significantly, a range of trade-offs is therefore possible between cost / complexity and the optimality of nonlinear distortion mitigation.

In another aspect, the present invention is an apparatus for mitigating non-channel nonlinear distortion of an optical signal carried on one of a plurality of wavelength channels in a wavelength division multiplexing (WDM) transmission system, comprising:
A first optical receiver configured to detect a measure of the total optical power of a plurality of wavelength channels;
A non-linear distortion compensator comprising means for applying phase modulation to an optical signal in proportion to the measure of total optical power is provided.

  In some embodiments, the non-linear distortion compensator includes an optical phase modulator having a modulation control input driven by a signal derived from the output of the first optical receiver, disposed in the optical transmission path of the optical signal. .

  The apparatus may advantageously further comprise a second optical receiver configured to detect the optical signal in order to perform nonlinear distortion compensation at the receiving end of the optical transmission link. In such an embodiment, the nonlinear distortion compensator is a signal derived from the output of the first optical receiver disposed in the electrical signal processing path of the received optical signal following detection by the second optical receiver. May include an electronic phase modulator having a modulation control input driven by.

  In some embodiments, the electronic phase modulator includes at least one analog phase modulator configured to apply phase modulation to the electrical signal output from the second optical receiver in accordance with the modulation control input.

In an alternative embodiment, the apparatus comprises at least one analog to digital converter (ADC) configured to convert the electrical signal output from the second optical receiver into a corresponding sequence of digital signal samples. ,
And a digital signal processor configured to apply phase modulation to the sequence of digital signal samples according to the modulation control input. More particularly, the apparatus may include yet another ADC configured to convert the output of the first optical receiver into a corresponding sequence of digital control samples, wherein the digital signal processor is a digital signal processor. It may be configured to apply phase modulation to the sequence of digital signal samples according to the sequence of control samples.

  The first optical receiver is preferably in accordance with the low pass characteristics, more preferably in accordance with the bandwidth that is narrower than the bandwidth of the optical signal and can be selected to maximize the compensation level of the interchannel nonlinear distortion of the optical signal. It is preferably configured to limit the bandwidth of the measure of total optical power. The first optical receiver typically has an essentially limited bandwidth, but the receiver may include a low pass filter and / or apply a control signal to the phase modulator. Further filtering of the output of the first optical receiver may be performed before In embodiments using digital signal processing, digital filtering of the digital control samples may be performed. Advantageously, this allows a highly flexible optimization of the mitigation of inter-channel nonlinear distortion, for example by using adaptive digital filters and / or by reconfiguring digital signal processing software.

  Further benefits, advantages, and preferred features of the method and apparatus of the present invention will become apparent in the description of the preferred embodiments below. The preferred embodiments should not be construed as limiting the invention as defined in any of the above descriptions or in the appended claims.

  Preferred embodiments of the present invention will be described with reference to the accompanying drawings.

FIG. 1 schematically illustrates a system for transmitting a signal over a non-linear optical channel including an apparatus for mitigation according to one embodiment of the present invention. 5 is a flowchart illustrating a method for mitigating non-channel nonlinear distortion of a received optical signal according to an embodiment of the present invention. FIG. 6 is a graph illustrating received signal quality as a function of channel number in a WDM transmission system, including results achieved in accordance with an embodiment of the present invention. FIG. 6 is a graph illustrating received signal quality as a function of optical power per channel in a WDM transmission system, including results achieved in accordance with one embodiment of the present invention.

Detailed Description of the Preferred Embodiment FIG. 1 schematically illustrates a system 100 for transmitting signals over a non-linear optical channel, according to an embodiment of the present invention. The system 100 includes a transmitter 102 and a receiving device 104 that implements the present invention. The transmitter 102 and the receiving device 104 are interconnected via a dispersion management optical channel 110. A booster amplifier 106 is provided at the transmitting end and includes a length of dispersion compensating fiber (DCF) 108 for precompensating for dispersion in the optical link 110. According to the conventional design, the booster amplifier 106 is a multi-stage amplifier such as a two-stage amplifier in which the DCF 108 is disposed between the first stage and the second stage. Such an arrangement is well known to those skilled in the field of optical communications and is therefore not described in detail herein.

  The optical link 110 includes a plurality of spans, each span including a standard single mode fiber (S-SMF) length 112 for transmission of optical signals and a multistage in-line amplifier 114 including a corresponding DCF length 116. In the example described herein, each S-SMF length 112 spans 95 km and 20 or 25 spans (ie, 1900 km or 2375 km in total) are used. However, it will be appreciated that these embodiments of optical channels are exemplary only and the present invention is applicable to any optical system that includes a non-linear optical channel for transmission of WDM signals.

  For convenience, only a single transmitter 102 is shown in the system 100. However, the system 100 is generally a WDM system, and in the example shown herein, eight wavelength channels are received by the receiver 104 having the received optical spectrum 118 schematically illustrated in FIG. Each of the eight wavelength channels has a bandwidth 120 (30 GHz in the example shown here), and each channel is separated by a spacing 122 (50 GHz in this example).

At the receiving end, the proportion of the received optical signal power is extracted using the optical tap 124 and directed to the first optical receiver 126. Receiver 126 includes a photodetector, such as photodiode 128, that is connected to associated electronic circuitry including amplifier 130 and then applies bandwidth limitations to the received photocurrent or voltage waveform. Is connected to a low-pass filter 132 that functions as follows. The output 134 of the first receiver 126 is a control signal that is a measure of the total optical power of the eight wavelength channels 118.

  The first optical receiver 126 is shown in FIG. 1 as consisting of three separate components: a photodetector 128, an amplifier 130, and a low pass filter 132, but this is to some extent. This is merely a convenient overview and in various embodiments, different component arrangements may be used. For example, it is understood that the photodetector 128 and the amplifier 130 each have an associated frequency response characteristic that, when combined with the characteristics of the low pass filter 132, determines the overall frequency response of the receiver 126. Will be done. Thus, achieving a particular desired overall frequency response involves considering the overall design of the receiver 126, and in some embodiments, depending on the design, all or part of the characteristics of the low pass filter 132. May be incorporated into the design of amplifier 130 and other associated electronic circuitry. Further, in some preferred embodiments, further digital processing of the control signal 134 may be performed, so that the desired bandwidth and / or frequency response characteristics of the receiver 126 are at least partially It may be achieved by digital signal processing techniques. It should be understood that all such variations are included within the scope of the present invention.

  Simultaneously with the detection of the total optical power by the receiver 126, individual transmitted optical signals, such as the optical signal generated by the transmitter 102, are separated from the WDM spectrum 118 by the WDM demultiplexer 136. One or more outputs of the WDM demultiplexer 136 are connected to corresponding optical receivers, and in particular, the exemplary system 100 includes a first one configured to detect the signal 138 transmitted by the transmitter 102. Two optical receivers 140 are shown. According to the illustrated embodiment, the receiver 140 has a plurality of outputs connected to corresponding analog / digital conversion and digital signal processing circuitry 142.

  More particularly, in a preferred embodiment of the present invention, as shown in FIG. 1, transmission system 100 encodes and modulates a digital signal for transmission on an optical channel to provide orthogonal frequency division multiplexing ( OFDM). For example, assuming a coherent transmission system, the optical OFDM signal is transmitted over the optical link 110 without a carrier wave. In order to maximize the available transmission capacity, the signal can be transmitted on two orthogonal polarization states of the light field. Accordingly, receiver 140 preferably includes a local oscillator (eg, a semiconductor or other laser light source) and an optical hybrid that separates both the quadrature polarization state and the in-phase and quadrature components of the transmitted coherent optical signal. Including a pair of balanced photodiodes for detection of received in-phase and quadrature components of two polarization multiplexed signals. Accordingly, the receiver 140 includes up to four electrical output ports. For simplicity, but without loss of generality, only a single polarization channel is utilized in the particular example described herein, so the receiver 140 has only two electrical output ports. These output ports are then directed to the digital conversion and processing block 142.

  Details on the generation and transmission of optical OFDM signals are disclosed in US patent application Ser. No. 12/085571 (also published as international application WO 2007/041799), the entire contents of which are incorporated herein by reference. Has been. This prior application also describes in detail the use of electronic digital signal processing techniques to compensate for residual linear dispersion effects in the optical link 110. In a preferred embodiment, distributed equalization is also performed in the digital processing block 142 of the system 100. However, the present invention is not limited to the combined use with an optical OFDM signal, but is expected to be effective for alleviating non-channel non-linear distortion in a system using another type of optical modulation, such as a system using coherent QPSK transmission. Is done.

  The schematic diagram of the system 100 includes several optical components and / or alternative components. In some embodiments of the present invention, the control output 134 is directed to the optical phase modulator 144 to apply phase modulation proportional to the amplitude of the control signal 134 to one or more of the received WDM signals. It may be done. In an alternative embodiment, the control signal output 134 is directed to an analog / digital converter 146 where it is converted to digital form and a corresponding sequence of digital control signal samples is directed to the digital signal processing block 142. Within the digital signal processor 142, a phase modulation proportional to the digital control signal samples is calculated and applied to the digital samples of the detected optical signal. In particular, in a system where the received optical signal (in-phase and quadrature components) is represented as a sequence of complex digital values, phase modulation can be achieved by converting control signal samples received from analog / digital converter 146 into unit magnitudes. And converting to a corresponding complex value having a phase proportional to the size of the control signal sample, and then multiplying the complex signal sample by the calculated phase modulation value.

  It should be noted that, at least in embodiments of the invention using an optical phase modulator 144, non-linear compensation can be performed without detection and / or processing of individual optical signals. Thus, although one embodiment 104 is described here where non-linear compensation is performed at the receiving end of the optical transmission link, one or more in the optical link or network, such as amplifier location or add / drop nodes. It will be appreciated that in-line compensation can also be performed at this point.

  Preferably, receiver device 104 also incorporates compensation for single channel nonlinear effects such as self phase modulation (SPM). The SPM compensation block 148 is shown in the schematic diagram of FIG. 1, and its operation follows the principles described in US patent application Ser. No. 12 / 445,386 (also published as international application WO2008 / 074085). Can be. This prior application also describes the use of single channel nonlinear pre-compensation, which can be implemented at the transmitter 102 according to a preferred embodiment of the present invention.

  The receiving device 104 may further include an optical bandpass filter 150, which allows the bandwidth of the WDM channel (such as the eight channels 118) to be more WDM received via the optical link 110. It is configured to select from among channels. More specifically, the interchannel nonlinear distortion experienced by a particular transmitted optical signal is most strongly generated by other WDM channels located at the closest frequency, so that the most affected WDM channel is generated by the first receiver 126. It is desirable to exclude WDM channels that are included in the generated control signal and that are more distant. In a WDM system having a large number of wavelength channels dispersed over a wide range of optical wavelengths, it is desirable to provide a plurality of receiving devices 104 each corresponding to a receiving channel of a specific band selected by the band pass filter 150. There is also. Depending on the configuration, the bandpass filter 150 may be a coarse WDM demultiplexer in which WDM channels of different bands are directed to each of a plurality of output ports. It may be desirable to provide overlap between the range of wavelengths covered by each band in order to maximize the degree to which interchannel nonlinear distortion can be mitigated in each individual optical signal.

  Further, the frequency response of the bandpass filter (or WDM demultiplexer) 150 includes a gradual “roll-off” at each edge of the frequency band, thereby causing the channel farthest away (in wavelength) from the intended received signal. It may be desirable to ensure that only a power proportion is directed to the first receiver 126. In this way, the more distant channel contributions to the control signal produced at the output 134 of the receiver 126 is balanced with the contribution of those channels to the interchannel nonlinear distortion of the optical signal of interest. In embodiment 100, the bandpass filter 150 is placed in front of the optical tap 124, but in an alternative embodiment, the bandpass filter 150 is placed between the optical tap 124 and the receiver 126 and is WDM. It is also possible to prevent the transmission WDM optical signal reaching the demultiplexer 136 from being affected by the characteristics of the band pass filter 150.

  Turning now to FIG. 2, a flowchart 200 illustrating a method for mitigating interchannel nonlinear distortion of a received optical signal is shown in accordance with a preferred embodiment of the present invention. The method shown in FIG. 2 is implemented by the receiving device 104 of the system 100 shown in FIG. In particular, at step 202, a band of a WDM channel is optionally selected corresponding to the function of the bandpass filter 150.

  At step 204, the instantaneous total optical power in the band of the WDM channel is detected, corresponding to the direction of the received WDM signal going to the first receiver 126 via the optical tap 124.

  At step 206, a bandwidth limit is applied to the detected signal. This step corresponds to the function of the low pass filter 132 and / or the appropriate digital signal processing performed by the processing block 142.

  In step 208, the resulting control signal is generated to produce a phase modulated signal (or corresponding sequence of digital control samples) that is proportional to the bandwidth limit measure of total optical power provided at the output 134 of the receiver 126. Is scaled, for example, by applying an appropriate “phase modulation factor”. For example, in analog implementations utilizing phase modulator 144, an appropriate modulation rate may be applied using an appropriate electrical gain and / or attenuation. This may be performed before or after the low pass filter 132 and may be incorporated into the low pass filter 132 and / or the electronic amplifier 130. In the digital implementation, the appropriate proportionality constant is easily implemented by numerical processing within the digital signal processing block 142.

  Finally, at step 210, the associated phase modulation is applied to the received optical signal, eg, by phase modulator 144 or by digital signal processing 142.

  With respect to bandwidth limitation 206, this is related to the effective bandwidth of the interchannel nonlinear distortion of the received optical signal that occurs during transmission over the optical link 110. In particular, the chromatic dispersion within individual S-SMF spans 112 and DCF compensator 116 limits the range of frequency components of the received optical signal that are most severely affected by interchannel nonlinear distortion. This is due to the relative “walk-off” of the WDM channel in the optical frequency domain, so that only relatively low frequency components are long enough to contribute significantly to the non-channel non-linear distortion. It remains phase matched over time. The maximum frequency component affected by interchannel distortion decreases within the level of increasing chromatic dispersion as the optical frequency difference between the contributing channels increases. Thus, the optimal bandwidth limitation depends on the dispersion map of the optical link 110 and the spacing of the WDM channels in the optical frequency domain. The optimal filter bandwidth can be estimated by appropriate calculations and / or by using experimental measurements or numerical simulations. For example, in a 50 GHz WDM channel grid, the most significant contribution to channel-to-channel nonlinear distortion occurs at less than 1 GHz, so this is a reduction in the bandwidth of the low-pass filter 132 in an exemplary system that utilizes a channel spacing of 50 GHz. Provide a good estimate. Numerical simulations performed by the inventors of the present invention show that the performance of the interchannel nonlinear distortion mitigation method is not critically dependent on achieving accurate optimization of the low pass filter bandwidth. showed that. Rather, there is a reasonable range of filter parameters that can achieve near optimal results. This allows the receiver device 104 to operate effectively across several optical channels in the WDM band 118 without requiring accurate optimization of the low pass filter 132 parameters. . If desired, filter parameter optimization may be performed adaptively within operating system 100 by adjusting the filter bandwidth to maximize the measure of received signal quality. As will be appreciated, adaptive optimization of filter parameters can be easily achieved if all or part of the bandwidth limitation is done in the digital domain.

  For transmission systems that utilize S-SMF (eg, 112) with respect to factors (ie, proportional) that relate the control signal generated at the output 134 of the receiver 126 to the level of phase modulation applied to the received optical signal. Such modulation generally includes a phase advance. The appropriate level of phase advance generally depends on the strength of the non-linear interaction between the WDM channels in the optical link 110. This further depends on the power level delivered to each span, the non-linear characteristics of the optical fibers (ie, S-SMF 112 and DCF 116) that make up these spans, and the non-linear effective length of each span. The concept of nonlinear effective length is well known in the art and accounts for the effect of damping on general nonlinear interactions. In particular, the signal is attenuated as it propagates through the span of the optical fiber, thus reducing the level of nonlinear interaction. That is, the nonlinear process becomes more prominent near the input end of each span where the optical power level is maximum. Therefore, the nonlinear effective length of a fiber span is generally slightly shorter than the actual physical length of the span.

  In practice, an approximate estimate of the appropriate phase modulation rate may be obtained based on the aforementioned considerations of the characteristics of the optical link 110. The resulting single constant can then be adjusted within the operating system 100 by an appropriate optimization process to maximize the appropriate measure of signal quality at the receivers 140, 142.

  Thus, in general, the amount of bandwidth limitation (applied in step 206) depends mainly on the dispersion characteristics of the optical link 110, and the phase modulation rate (applied in step 208) mainly depends on, for example, fiber Depends on factors that contribute to the strength of the nonlinear interaction, such as nonlinearity and attenuation of light, and light irradiation power. Thus, these two parameters of the receiving device 104 are relatively independent of each other and can therefore be easily optimized by an independent optimization process. Furthermore, an online optimization process can be easily implemented because each represents a single maximum value of the corresponding received signal quality within an easily identified range.

  In order to evaluate the effectiveness of embodiments of the present invention, several computer simulations of a transmission system, approximately consistent with the exemplary system 100, were performed. For simplicity, these simulations were applied to a model system consisting of a plurality of optical fiber transmission spans, each assumed to be the same. The simulated system includes 8 WDM channels, each channel having a single 30 GHz bandwidth light with 1024 subcarriers modulated according to a 4QAM scheme resulting in a raw data rate of 60 Gbit / s in a single polarization. Carries OFDM signals. The WDM channel spacing is 50 GHz.

  For the simulated system optical dispersion map, initial chromatic dispersion pre-compensation (ie compensation through DCF 108) was applied, for a total of 1530 ps / nm. Each span includes a 95 km S-SMF 112 followed by an amplifier 114 incorporating a DCF 116 that is configured to compensate the dispersion in the S-SMF 112 by 85 ps / nm. The remaining dispersion is compensated at the receiver. Two-stage amplifiers 106, 114 compensate for the loss in each span and have a noise figure of 5 dB. The amplifier gain is configured such that the power output from each DCF length 116 is the same as the power at the output of each S-SMF length 112, minimizing nonlinear effects in the DCF 116.

  In various simulations whose results are described in detail below with reference to FIGS. 3 and 4, the total number of spans 110 is different (either 20 or 25 spans) and sent to each span. The power is also different, from -10 dBm to 0 dBm per channel.

  With respect to the simulated receiver 104, the interchannel nonlinearity compensator consists of a first receiver 126 and a phase modulator 144 and is placed in front of the demultiplexer 136 to simultaneously compensate all eight WDM channels. Is possible. As can be seen from the foregoing description, referring to FIGS. 1 and 2, this configuration is generally not expected to provide optimal compensation for all eight WDM channels, but simulation results are shown using this technique. It can be shown that inter-channel nonlinear distortion can be mitigated in a plurality of WDM channels, and that the simultaneous compensation of the plurality of WDM channels can advantageously reduce the complexity and / or cost of the receiver 104. ing. In these simulations, the characteristics of the low-pass filter 132 are changed to the fourth WDM channel (ie, the band) by systematically changing the filter passband and transition band widths to find the maximum compensation level of the interchannel nonlinear distortion. The numerical optimization was performed so that it would fit in the vicinity of the center.

  The eight WDM channels are demultiplexed by demultiplexer 136, individually SPM compensated by SPM compensator 148, and detected using coherent receiver 140. The 4QAM modulated data is then recovered by digital processing 142 of the received signal, and the resulting electrical signal quality Q is determined. In these particular examples, the signal quality Q is defined as the square of the average distance of 4QAM symbols from the relevant axis of the complex plane divided by the corresponding symbol variance. That is, Q is a measure of the likelihood (or ratio) of detection errors, and a higher Q value represents a better quality signal. Significantly, the value of Q has a lower bound slightly larger than 9 dB, and above 9 dB, error-free transmission can be achieved by using an appropriate forward error correction (FEC) algorithm. Thus, this Q value constitutes the “FEC limit” for error-free transmission.

  FIG. 3 is a graph 300 showing the average value Q (on axis 302) for each of the eight WDM channels (on axis 304). The graph 300 also shows the FEC limit 306. The transmission system includes 20 spans of 95 km each over a total of 1900 km and has a light output power of -2 dBm per channel. A circle 308 represents the received signal quality without compensation. Triangle 310 represents received signal quality with only non-channel non-linear distortion mitigation. Square 312 represents the received signal quality with only single channel (ie, SPM) nonlinear distortion compensation. The diamond represents the received signal quality with non-linear distortion mitigation both between channels and a single channel. It is clear that without nonlinear distortion relaxation, the central channel (eg channel 3 to channel 6) exhibits the lowest performance due to strong inter-channel effects from the other channels. The relaxation of the individual channel SPM gives the channel maximum benefit at the end of the band, eg channel 1 or channel 8. This is because these channels are least affected by non-channel non-linear distortion. Conversely, mitigation of only the interchannel effect provides the greatest benefit to the central channel, eg, channel 3 to channel 6. When both single channel and channel-to-channel nonlinear distortion relaxation is performed, the quality of all eight resulting WDM channels exceeds the FEC limit.

  Turning now to FIG. 4, a graph 400 is shown in which signal quality (on axis 402) is plotted against light output power per channel (on axis 404). Curves 406, 408 represent systems where no compensation for nonlinear distortion is applied. Curves 410, 412 represent a system to which both single channel and interchannel nonlinear strain relaxation is applied, according to embodiments of the present invention. Further, curves 406 and 410 represent a system including 20 transmission spans of 95 km each over a total of 1900 km, and curves 408 and 412 represent a system including 25 spans of 95 km over a total transmission of 2375 km. Four curves 406, 408, 410, 412 all show the signal quality for the worst WDM channel in each case.

  From the results shown in graph 400, for any transmission distance, using nonlinearity compensation including inter-channel mitigation according to embodiments of the present invention results in a corresponding minimum (ie, worst case) increase in received signal quality. It is clear that it is possible to use a higher optimal light output power per channel with In particular, without nonlinearity mitigation, the optimum optical power is around -6 dBm (414), and with nonlinearity mitigation, this increases by approximately 2 dB (416), which is 1.3 dB of the worst received signal quality. Improvement (418) is achieved.

  Furthermore, when nonlinearity mitigation is used, it is possible to exceed the FEC limit in all WDM channels, even over a distance of 2375 km over 25 dispersion management spans, but on the same transmission link without nonlinearity compensation. This is impossible.

  In addition, referring to the results shown in graph 300 of FIG. 3, the worst performing channel represented by curves 406, 408 of graph 400 is the channel located near the center of the WDM band when no compensation is applied. It is clear. A significant improvement in the quality of these channels sufficient to exceed the FEC limit can be achieved only through the use of interchannel nonlinear distortion relaxation according to embodiments of the present invention. (In comparison, the channels at the end of the WDM band, eg, channel 1 and channel 8 achieve a majority of the complete signal quality improvement only by SPM compensation.) Therefore, only by using the embodiments of the present invention. Clearly, a significant improvement in received signal quality can be achieved across all channels of the WDM transmission system. In particular, according to various embodiments disclosed herein, advantageously, a low bandwidth receiver 126 is used to obtain a single measure of total optical power in multiple wavelength channels. It is possible to achieve the necessary improvements using simple implementations, and then advantageously using this measure, reception in one or more of the wavelength channels simultaneously or per channel The influence of non-channel nonlinear distortion of the optical signal can be mitigated.

  While several implementation options and variations have been described, further modifications will be apparent to those skilled in the art. Accordingly, the invention is not to be limited to the specific embodiments described herein, but the scope of which is defined by the appended claims.

Claims (17)

A method for mitigating non-channel nonlinear distortion of an optical signal carried on one of a plurality of wavelength channels in a wavelength division multiplexing (WDM) transmission system, comprising:
A step that generates a control signal which is a bandwidth limitation measure of the instantaneous total optical power of said plurality of wavelength channels,
Demultiplexing individual signals from multiple wavelength channels;
And applying a phase modulation to in proportion to the control signal, the individual signals,
Including methods. The method of claim 1, wherein a bandwidth of a measure of instantaneous optical power is limited according to a low pass characteristic having a narrower bandwidth than the bandwidth of the individual signals. The method according to claim 2, further comprising selecting or optimizing the low-pass characteristic so as to maximize a compensation level of inter-channel nonlinear distortion of the individual signals.   The method of claim 1, wherein the applied phase modulation is advancing phase. The method of claim 1, wherein a proportionality constant between the control signal and the level of phase modulation is determined according to a measure of an effective nonlinear length of the WDM transmission system.   The method of claim 1, wherein the plurality of wavelength channels comprises a band of channels selected from a larger number of transmission WDM channels. The method of claim 1, wherein the individual signals are carried at wavelengths located near a central wavelength of the plurality of wavelength channels. The phase modulation is applied to individual signals carried on one or more corresponding ones of the wavelength channels in the WDM transmission system , or separately to one or more signals. Method. An apparatus for mitigating nonlinear distortion between channels of an optical signal carried on one of a plurality of wavelength channels in a wavelength division multiplexing (WDM) transmission system,
Detecting a plurality of wavelengths channels, and a first optical receiver for generating a control signal, which is configured such that the bandwidth limiting measure of the instantaneous total optical power of said plurality of wavelength channels,
A demultiplexer that demultiplexes individual signals from multiple wavelength channels;
A nonlinear distortion compensator including means for applying phase modulation to the individual signals in proportion to the control signal ;
Including the device. The nonlinear distortion compensator previously includes a phase modulator disposed in the optical transmission line of the optical signal,
The apparatus of claim 9, wherein the phase modulator has a modulation control input driven by the control signal . A second optical receiver configured to detect the individual signals;
The non-linear distortion compensator includes an electronic phase modulator disposed in an electrical signal processing path of the received optical signal following detection by the second optical receiver;
The apparatus of claim 9, wherein the phase modulator has a modulation control input driven by the control signal .   The electronic phase modulator includes at least one analog phase modulator configured to apply phase modulation to an electrical signal output from the second optical receiver according to the modulation control input. Equipment. At least one analog / digital converter (ADC) configured to convert the electrical signal output from the second optical receiver into a corresponding sequence of digital signal samples;
A digital signal processor configured to apply phase modulation to the sequence of digital signal samples according to the modulation control input;
The apparatus of claim 11 further comprising: Comprising another ADC configured to convert the output of the first optical receiver into a corresponding sequence of digital control samples;
The apparatus of claim 13, wherein the digital signal processor is configured to apply phase modulation to the sequence of digital signal samples according to the sequence of digital control samples.   The apparatus of claim 9, wherein the first optical receiver is configured to limit a bandwidth of the measure of the total optical power according to low pass characteristics. The apparatus of claim 15, wherein the low pass characteristic is selected to maximize a compensation level of inter-channel nonlinear distortion of the individual signals.   The digital signal processor is further configured to digitally filter the digital control samples having low pass characteristics selected to maximize a compensation level of interchannel nonlinear distortion of the optical signal. The apparatus according to 16.

Images